Commit | Line | Data |
---|---|---|
920dae64 AT |
1 | # Pod::PlainText -- Convert POD data to formatted ASCII text. |
2 | # $Id: Text.pm,v 2.1 1999/09/20 11:53:33 eagle Exp $ | |
3 | # | |
4 | # Copyright 1999-2000 by Russ Allbery <rra@stanford.edu> | |
5 | # | |
6 | # This program is free software; you can redistribute it and/or modify it | |
7 | # under the same terms as Perl itself. | |
8 | # | |
9 | # This module is intended to be a replacement for Pod::Text, and attempts to | |
10 | # match its output except for some specific circumstances where other | |
11 | # decisions seemed to produce better output. It uses Pod::Parser and is | |
12 | # designed to be very easy to subclass. | |
13 | ||
14 | ############################################################################ | |
15 | # Modules and declarations | |
16 | ############################################################################ | |
17 | ||
18 | package Pod::PlainText; | |
19 | ||
20 | require 5.005; | |
21 | ||
22 | use Carp qw(carp croak); | |
23 | use Pod::Select (); | |
24 | ||
25 | use strict; | |
26 | use vars qw(@ISA %ESCAPES $VERSION); | |
27 | ||
28 | # We inherit from Pod::Select instead of Pod::Parser so that we can be used | |
29 | # by Pod::Usage. | |
30 | @ISA = qw(Pod::Select); | |
31 | ||
32 | $VERSION = '2.02'; | |
33 | ||
34 | ||
35 | ############################################################################ | |
36 | # Table of supported E<> escapes | |
37 | ############################################################################ | |
38 | ||
39 | # This table is taken near verbatim from Pod::PlainText in Pod::Parser, | |
40 | # which got it near verbatim from the original Pod::Text. It is therefore | |
41 | # credited to Tom Christiansen, and I'm glad I didn't have to write it. :) | |
42 | %ESCAPES = ( | |
43 | 'amp' => '&', # ampersand | |
44 | 'lt' => '<', # left chevron, less-than | |
45 | 'gt' => '>', # right chevron, greater-than | |
46 | 'quot' => '"', # double quote | |
47 | ||
48 | "Aacute" => "\xC1", # capital A, acute accent | |
49 | "aacute" => "\xE1", # small a, acute accent | |
50 | "Acirc" => "\xC2", # capital A, circumflex accent | |
51 | "acirc" => "\xE2", # small a, circumflex accent | |
52 | "AElig" => "\xC6", # capital AE diphthong (ligature) | |
53 | "aelig" => "\xE6", # small ae diphthong (ligature) | |
54 | "Agrave" => "\xC0", # capital A, grave accent | |
55 | "agrave" => "\xE0", # small a, grave accent | |
56 | "Aring" => "\xC5", # capital A, ring | |
57 | "aring" => "\xE5", # small a, ring | |
58 | "Atilde" => "\xC3", # capital A, tilde | |
59 | "atilde" => "\xE3", # small a, tilde | |
60 | "Auml" => "\xC4", # capital A, dieresis or umlaut mark | |
61 | "auml" => "\xE4", # small a, dieresis or umlaut mark | |
62 | "Ccedil" => "\xC7", # capital C, cedilla | |
63 | "ccedil" => "\xE7", # small c, cedilla | |
64 | "Eacute" => "\xC9", # capital E, acute accent | |
65 | "eacute" => "\xE9", # small e, acute accent | |
66 | "Ecirc" => "\xCA", # capital E, circumflex accent | |
67 | "ecirc" => "\xEA", # small e, circumflex accent | |
68 | "Egrave" => "\xC8", # capital E, grave accent | |
69 | "egrave" => "\xE8", # small e, grave accent | |
70 | "ETH" => "\xD0", # capital Eth, Icelandic | |
71 | "eth" => "\xF0", # small eth, Icelandic | |
72 | "Euml" => "\xCB", # capital E, dieresis or umlaut mark | |
73 | "euml" => "\xEB", # small e, dieresis or umlaut mark | |
74 | "Iacute" => "\xCD", # capital I, acute accent | |
75 | "iacute" => "\xED", # small i, acute accent | |
76 | "Icirc" => "\xCE", # capital I, circumflex accent | |
77 | "icirc" => "\xEE", # small i, circumflex accent | |
78 | "Igrave" => "\xCD", # capital I, grave accent | |
79 | "igrave" => "\xED", # small i, grave accent | |
80 | "Iuml" => "\xCF", # capital I, dieresis or umlaut mark | |
81 | "iuml" => "\xEF", # small i, dieresis or umlaut mark | |
82 | "Ntilde" => "\xD1", # capital N, tilde | |
83 | "ntilde" => "\xF1", # small n, tilde | |
84 | "Oacute" => "\xD3", # capital O, acute accent | |
85 | "oacute" => "\xF3", # small o, acute accent | |
86 | "Ocirc" => "\xD4", # capital O, circumflex accent | |
87 | "ocirc" => "\xF4", # small o, circumflex accent | |
88 | "Ograve" => "\xD2", # capital O, grave accent | |
89 | "ograve" => "\xF2", # small o, grave accent | |
90 | "Oslash" => "\xD8", # capital O, slash | |
91 | "oslash" => "\xF8", # small o, slash | |
92 | "Otilde" => "\xD5", # capital O, tilde | |
93 | "otilde" => "\xF5", # small o, tilde | |
94 | "Ouml" => "\xD6", # capital O, dieresis or umlaut mark | |
95 | "ouml" => "\xF6", # small o, dieresis or umlaut mark | |
96 | "szlig" => "\xDF", # small sharp s, German (sz ligature) | |
97 | "THORN" => "\xDE", # capital THORN, Icelandic | |
98 | "thorn" => "\xFE", # small thorn, Icelandic | |
99 | "Uacute" => "\xDA", # capital U, acute accent | |
100 | "uacute" => "\xFA", # small u, acute accent | |
101 | "Ucirc" => "\xDB", # capital U, circumflex accent | |
102 | "ucirc" => "\xFB", # small u, circumflex accent | |
103 | "Ugrave" => "\xD9", # capital U, grave accent | |
104 | "ugrave" => "\xF9", # small u, grave accent | |
105 | "Uuml" => "\xDC", # capital U, dieresis or umlaut mark | |
106 | "uuml" => "\xFC", # small u, dieresis or umlaut mark | |
107 | "Yacute" => "\xDD", # capital Y, acute accent | |
108 | "yacute" => "\xFD", # small y, acute accent | |
109 | "yuml" => "\xFF", # small y, dieresis or umlaut mark | |
110 | ||
111 | "lchevron" => "\xAB", # left chevron (double less than) | |
112 | "rchevron" => "\xBB", # right chevron (double greater than) | |
113 | ); | |
114 | ||
115 | ||
116 | ############################################################################ | |
117 | # Initialization | |
118 | ############################################################################ | |
119 | ||
120 | # Initialize the object. Must be sure to call our parent initializer. | |
121 | sub initialize { | |
122 | my $self = shift; | |
123 | ||
124 | $$self{alt} = 0 unless defined $$self{alt}; | |
125 | $$self{indent} = 4 unless defined $$self{indent}; | |
126 | $$self{loose} = 0 unless defined $$self{loose}; | |
127 | $$self{sentence} = 0 unless defined $$self{sentence}; | |
128 | $$self{width} = 76 unless defined $$self{width}; | |
129 | ||
130 | $$self{INDENTS} = []; # Stack of indentations. | |
131 | $$self{MARGIN} = $$self{indent}; # Current left margin in spaces. | |
132 | ||
133 | $self->SUPER::initialize; | |
134 | } | |
135 | ||
136 | ||
137 | ############################################################################ | |
138 | # Core overrides | |
139 | ############################################################################ | |
140 | ||
141 | # Called for each command paragraph. Gets the command, the associated | |
142 | # paragraph, the line number, and a Pod::Paragraph object. Just dispatches | |
143 | # the command to a method named the same as the command. =cut is handled | |
144 | # internally by Pod::Parser. | |
145 | sub command { | |
146 | my $self = shift; | |
147 | my $command = shift; | |
148 | return if $command eq 'pod'; | |
149 | return if ($$self{EXCLUDE} && $command ne 'end'); | |
150 | $self->item ("\n") if defined $$self{ITEM}; | |
151 | $command = 'cmd_' . $command; | |
152 | $self->$command (@_); | |
153 | } | |
154 | ||
155 | # Called for a verbatim paragraph. Gets the paragraph, the line number, and | |
156 | # a Pod::Paragraph object. Just output it verbatim, but with tabs converted | |
157 | # to spaces. | |
158 | sub verbatim { | |
159 | my $self = shift; | |
160 | return if $$self{EXCLUDE}; | |
161 | $self->item if defined $$self{ITEM}; | |
162 | local $_ = shift; | |
163 | return if /^\s*$/; | |
164 | s/^(\s*\S+)/(' ' x $$self{MARGIN}) . $1/gme; | |
165 | $self->output ($_); | |
166 | } | |
167 | ||
168 | # Called for a regular text block. Gets the paragraph, the line number, and | |
169 | # a Pod::Paragraph object. Perform interpolation and output the results. | |
170 | sub textblock { | |
171 | my $self = shift; | |
172 | return if $$self{EXCLUDE}; | |
173 | $self->output ($_[0]), return if $$self{VERBATIM}; | |
174 | local $_ = shift; | |
175 | my $line = shift; | |
176 | ||
177 | # Perform a little magic to collapse multiple L<> references. This is | |
178 | # here mostly for backwards-compatibility. We'll just rewrite the whole | |
179 | # thing into actual text at this part, bypassing the whole internal | |
180 | # sequence parsing thing. | |
181 | s{ | |
182 | ( | |
183 | L< # A link of the form L</something>. | |
184 | / | |
185 | ( | |
186 | [:\w]+ # The item has to be a simple word... | |
187 | (\(\))? # ...or simple function. | |
188 | ) | |
189 | > | |
190 | ( | |
191 | ,?\s+(and\s+)? # Allow lots of them, conjuncted. | |
192 | L< | |
193 | / | |
194 | ( | |
195 | [:\w]+ | |
196 | (\(\))? | |
197 | ) | |
198 | > | |
199 | )+ | |
200 | ) | |
201 | } { | |
202 | local $_ = $1; | |
203 | s%L</([^>]+)>%$1%g; | |
204 | my @items = split /(?:,?\s+(?:and\s+)?)/; | |
205 | my $string = "the "; | |
206 | my $i; | |
207 | for ($i = 0; $i < @items; $i++) { | |
208 | $string .= $items[$i]; | |
209 | $string .= ", " if @items > 2 && $i != $#items; | |
210 | $string .= " and " if ($i == $#items - 1); | |
211 | } | |
212 | $string .= " entries elsewhere in this document"; | |
213 | $string; | |
214 | }gex; | |
215 | ||
216 | # Now actually interpolate and output the paragraph. | |
217 | $_ = $self->interpolate ($_, $line); | |
218 | s/\s+$/\n/; | |
219 | if (defined $$self{ITEM}) { | |
220 | $self->item ($_ . "\n"); | |
221 | } else { | |
222 | $self->output ($self->reformat ($_ . "\n")); | |
223 | } | |
224 | } | |
225 | ||
226 | # Called for an interior sequence. Gets the command, argument, and a | |
227 | # Pod::InteriorSequence object and is expected to return the resulting text. | |
228 | # Calls code, bold, italic, file, and link to handle those types of | |
229 | # sequences, and handles S<>, E<>, X<>, and Z<> directly. | |
230 | sub interior_sequence { | |
231 | my $self = shift; | |
232 | my $command = shift; | |
233 | local $_ = shift; | |
234 | return '' if ($command eq 'X' || $command eq 'Z'); | |
235 | ||
236 | # Expand escapes into the actual character now, carping if invalid. | |
237 | if ($command eq 'E') { | |
238 | return $ESCAPES{$_} if defined $ESCAPES{$_}; | |
239 | carp "Unknown escape: E<$_>"; | |
240 | return "E<$_>"; | |
241 | } | |
242 | ||
243 | # For all the other sequences, empty content produces no output. | |
244 | return if $_ eq ''; | |
245 | ||
246 | # For S<>, compress all internal whitespace and then map spaces to \01. | |
247 | # When we output the text, we'll map this back. | |
248 | if ($command eq 'S') { | |
249 | s/\s{2,}/ /g; | |
250 | tr/ /\01/; | |
251 | return $_; | |
252 | } | |
253 | ||
254 | # Anything else needs to get dispatched to another method. | |
255 | if ($command eq 'B') { return $self->seq_b ($_) } | |
256 | elsif ($command eq 'C') { return $self->seq_c ($_) } | |
257 | elsif ($command eq 'F') { return $self->seq_f ($_) } | |
258 | elsif ($command eq 'I') { return $self->seq_i ($_) } | |
259 | elsif ($command eq 'L') { return $self->seq_l ($_) } | |
260 | else { carp "Unknown sequence $command<$_>" } | |
261 | } | |
262 | ||
263 | # Called for each paragraph that's actually part of the POD. We take | |
264 | # advantage of this opportunity to untabify the input. | |
265 | sub preprocess_paragraph { | |
266 | my $self = shift; | |
267 | local $_ = shift; | |
268 | 1 while s/^(.*?)(\t+)/$1 . ' ' x (length ($2) * 8 - length ($1) % 8)/me; | |
269 | $_; | |
270 | } | |
271 | ||
272 | ||
273 | ############################################################################ | |
274 | # Command paragraphs | |
275 | ############################################################################ | |
276 | ||
277 | # All command paragraphs take the paragraph and the line number. | |
278 | ||
279 | # First level heading. | |
280 | sub cmd_head1 { | |
281 | my $self = shift; | |
282 | local $_ = shift; | |
283 | s/\s+$//; | |
284 | $_ = $self->interpolate ($_, shift); | |
285 | if ($$self{alt}) { | |
286 | $self->output ("\n==== $_ ====\n\n"); | |
287 | } else { | |
288 | $_ .= "\n" if $$self{loose}; | |
289 | $self->output ($_ . "\n"); | |
290 | } | |
291 | } | |
292 | ||
293 | # Second level heading. | |
294 | sub cmd_head2 { | |
295 | my $self = shift; | |
296 | local $_ = shift; | |
297 | s/\s+$//; | |
298 | $_ = $self->interpolate ($_, shift); | |
299 | if ($$self{alt}) { | |
300 | $self->output ("\n== $_ ==\n\n"); | |
301 | } else { | |
302 | $self->output (' ' x ($$self{indent} / 2) . $_ . "\n\n"); | |
303 | } | |
304 | } | |
305 | ||
306 | # Start a list. | |
307 | sub cmd_over { | |
308 | my $self = shift; | |
309 | local $_ = shift; | |
310 | unless (/^[-+]?\d+\s+$/) { $_ = $$self{indent} } | |
311 | push (@{ $$self{INDENTS} }, $$self{MARGIN}); | |
312 | $$self{MARGIN} += ($_ + 0); | |
313 | } | |
314 | ||
315 | # End a list. | |
316 | sub cmd_back { | |
317 | my $self = shift; | |
318 | $$self{MARGIN} = pop @{ $$self{INDENTS} }; | |
319 | unless (defined $$self{MARGIN}) { | |
320 | carp "Unmatched =back"; | |
321 | $$self{MARGIN} = $$self{indent}; | |
322 | } | |
323 | } | |
324 | ||
325 | # An individual list item. | |
326 | sub cmd_item { | |
327 | my $self = shift; | |
328 | if (defined $$self{ITEM}) { $self->item } | |
329 | local $_ = shift; | |
330 | s/\s+$//; | |
331 | $$self{ITEM} = $self->interpolate ($_); | |
332 | } | |
333 | ||
334 | # Begin a block for a particular translator. Setting VERBATIM triggers | |
335 | # special handling in textblock(). | |
336 | sub cmd_begin { | |
337 | my $self = shift; | |
338 | local $_ = shift; | |
339 | my ($kind) = /^(\S+)/ or return; | |
340 | if ($kind eq 'text') { | |
341 | $$self{VERBATIM} = 1; | |
342 | } else { | |
343 | $$self{EXCLUDE} = 1; | |
344 | } | |
345 | } | |
346 | ||
347 | # End a block for a particular translator. We assume that all =begin/=end | |
348 | # pairs are properly closed. | |
349 | sub cmd_end { | |
350 | my $self = shift; | |
351 | $$self{EXCLUDE} = 0; | |
352 | $$self{VERBATIM} = 0; | |
353 | } | |
354 | ||
355 | # One paragraph for a particular translator. Ignore it unless it's intended | |
356 | # for text, in which case we treat it as a verbatim text block. | |
357 | sub cmd_for { | |
358 | my $self = shift; | |
359 | local $_ = shift; | |
360 | my $line = shift; | |
361 | return unless s/^text\b[ \t]*\n?//; | |
362 | $self->verbatim ($_, $line); | |
363 | } | |
364 | ||
365 | ||
366 | ############################################################################ | |
367 | # Interior sequences | |
368 | ############################################################################ | |
369 | ||
370 | # The simple formatting ones. These are here mostly so that subclasses can | |
371 | # override them and do more complicated things. | |
372 | sub seq_b { return $_[0]{alt} ? "``$_[1]''" : $_[1] } | |
373 | sub seq_c { return $_[0]{alt} ? "``$_[1]''" : "`$_[1]'" } | |
374 | sub seq_f { return $_[0]{alt} ? "\"$_[1]\"" : $_[1] } | |
375 | sub seq_i { return '*' . $_[1] . '*' } | |
376 | ||
377 | # The complicated one. Handle links. Since this is plain text, we can't | |
378 | # actually make any real links, so this is all to figure out what text we | |
379 | # print out. | |
380 | sub seq_l { | |
381 | my $self = shift; | |
382 | local $_ = shift; | |
383 | ||
384 | # Smash whitespace in case we were split across multiple lines. | |
385 | s/\s+/ /g; | |
386 | ||
387 | # If we were given any explicit text, just output it. | |
388 | if (/^([^|]+)\|/) { return $1 } | |
389 | ||
390 | # Okay, leading and trailing whitespace isn't important; get rid of it. | |
391 | s/^\s+//; | |
392 | s/\s+$//; | |
393 | ||
394 | # Default to using the whole content of the link entry as a section | |
395 | # name. Note that L<manpage/> forces a manpage interpretation, as does | |
396 | # something looking like L<manpage(section)>. The latter is an | |
397 | # enhancement over the original Pod::Text. | |
398 | my ($manpage, $section) = ('', $_); | |
399 | if (/^(?:https?|ftp|news):/) { | |
400 | # a URL | |
401 | return $_; | |
402 | } elsif (/^"\s*(.*?)\s*"$/) { | |
403 | $section = '"' . $1 . '"'; | |
404 | } elsif (m/^[-:.\w]+(?:\(\S+\))?$/) { | |
405 | ($manpage, $section) = ($_, ''); | |
406 | } elsif (m%/%) { | |
407 | ($manpage, $section) = split (/\s*\/\s*/, $_, 2); | |
408 | } | |
409 | ||
410 | my $text = ''; | |
411 | # Now build the actual output text. | |
412 | if (!length $section) { | |
413 | $text = "the $manpage manpage" if length $manpage; | |
414 | } elsif ($section =~ /^[:\w]+(?:\(\))?/) { | |
415 | $text .= 'the ' . $section . ' entry'; | |
416 | $text .= (length $manpage) ? " in the $manpage manpage" | |
417 | : " elsewhere in this document"; | |
418 | } else { | |
419 | $section =~ s/^\"\s*//; | |
420 | $section =~ s/\s*\"$//; | |
421 | $text .= 'the section on "' . $section . '"'; | |
422 | $text .= " in the $manpage manpage" if length $manpage; | |
423 | } | |
424 | $text; | |
425 | } | |
426 | ||
427 | ||
428 | ############################################################################ | |
429 | # List handling | |
430 | ############################################################################ | |
431 | ||
432 | # This method is called whenever an =item command is complete (in other | |
433 | # words, we've seen its associated paragraph or know for certain that it | |
434 | # doesn't have one). It gets the paragraph associated with the item as an | |
435 | # argument. If that argument is empty, just output the item tag; if it | |
436 | # contains a newline, output the item tag followed by the newline. | |
437 | # Otherwise, see if there's enough room for us to output the item tag in the | |
438 | # margin of the text or if we have to put it on a separate line. | |
439 | sub item { | |
440 | my $self = shift; | |
441 | local $_ = shift; | |
442 | my $tag = $$self{ITEM}; | |
443 | unless (defined $tag) { | |
444 | carp "item called without tag"; | |
445 | return; | |
446 | } | |
447 | undef $$self{ITEM}; | |
448 | my $indent = $$self{INDENTS}[-1]; | |
449 | unless (defined $indent) { $indent = $$self{indent} } | |
450 | my $space = ' ' x $indent; | |
451 | $space =~ s/^ /:/ if $$self{alt}; | |
452 | if (!$_ || /^\s+$/ || ($$self{MARGIN} - $indent < length ($tag) + 1)) { | |
453 | my $margin = $$self{MARGIN}; | |
454 | $$self{MARGIN} = $indent; | |
455 | my $output = $self->reformat ($tag); | |
456 | $output =~ s/\n*$/\n/; | |
457 | $self->output ($output); | |
458 | $$self{MARGIN} = $margin; | |
459 | $self->output ($self->reformat ($_)) if /\S/; | |
460 | } else { | |
461 | $_ = $self->reformat ($_); | |
462 | s/^ /:/ if ($$self{alt} && $indent > 0); | |
463 | my $tagspace = ' ' x length $tag; | |
464 | s/^($space)$tagspace/$1$tag/ or warn "Bizarre space in item"; | |
465 | $self->output ($_); | |
466 | } | |
467 | } | |
468 | ||
469 | ||
470 | ############################################################################ | |
471 | # Output formatting | |
472 | ############################################################################ | |
473 | ||
474 | # Wrap a line, indenting by the current left margin. We can't use | |
475 | # Text::Wrap because it plays games with tabs. We can't use formline, even | |
476 | # though we'd really like to, because it screws up non-printing characters. | |
477 | # So we have to do the wrapping ourselves. | |
478 | sub wrap { | |
479 | my $self = shift; | |
480 | local $_ = shift; | |
481 | my $output = ''; | |
482 | my $spaces = ' ' x $$self{MARGIN}; | |
483 | my $width = $$self{width} - $$self{MARGIN}; | |
484 | while (length > $width) { | |
485 | if (s/^([^\n]{0,$width})\s+// || s/^([^\n]{$width})//) { | |
486 | $output .= $spaces . $1 . "\n"; | |
487 | } else { | |
488 | last; | |
489 | } | |
490 | } | |
491 | $output .= $spaces . $_; | |
492 | $output =~ s/\s+$/\n\n/; | |
493 | $output; | |
494 | } | |
495 | ||
496 | # Reformat a paragraph of text for the current margin. Takes the text to | |
497 | # reformat and returns the formatted text. | |
498 | sub reformat { | |
499 | my $self = shift; | |
500 | local $_ = shift; | |
501 | ||
502 | # If we're trying to preserve two spaces after sentences, do some | |
503 | # munging to support that. Otherwise, smash all repeated whitespace. | |
504 | if ($$self{sentence}) { | |
505 | s/ +$//mg; | |
506 | s/\.\n/. \n/g; | |
507 | s/\n/ /g; | |
508 | s/ +/ /g; | |
509 | } else { | |
510 | s/\s+/ /g; | |
511 | } | |
512 | $self->wrap ($_); | |
513 | } | |
514 | ||
515 | # Output text to the output device. | |
516 | sub output { $_[1] =~ tr/\01/ /; print { $_[0]->output_handle } $_[1] } | |
517 | ||
518 | ||
519 | ############################################################################ | |
520 | # Backwards compatibility | |
521 | ############################################################################ | |
522 | ||
523 | # The old Pod::Text module did everything in a pod2text() function. This | |
524 | # tries to provide the same interface for legacy applications. | |
525 | sub pod2text { | |
526 | my @args; | |
527 | ||
528 | # This is really ugly; I hate doing option parsing in the middle of a | |
529 | # module. But the old Pod::Text module supported passing flags to its | |
530 | # entry function, so handle -a and -<number>. | |
531 | while ($_[0] =~ /^-/) { | |
532 | my $flag = shift; | |
533 | if ($flag eq '-a') { push (@args, alt => 1) } | |
534 | elsif ($flag =~ /^-(\d+)$/) { push (@args, width => $1) } | |
535 | else { | |
536 | unshift (@_, $flag); | |
537 | last; | |
538 | } | |
539 | } | |
540 | ||
541 | # Now that we know what arguments we're using, create the parser. | |
542 | my $parser = Pod::PlainText->new (@args); | |
543 | ||
544 | # If two arguments were given, the second argument is going to be a file | |
545 | # handle. That means we want to call parse_from_filehandle(), which | |
546 | # means we need to turn the first argument into a file handle. Magic | |
547 | # open will handle the <&STDIN case automagically. | |
548 | if (defined $_[1]) { | |
549 | local *IN; | |
550 | unless (open (IN, $_[0])) { | |
551 | croak ("Can't open $_[0] for reading: $!\n"); | |
552 | return; | |
553 | } | |
554 | $_[0] = \*IN; | |
555 | return $parser->parse_from_filehandle (@_); | |
556 | } else { | |
557 | return $parser->parse_from_file (@_); | |
558 | } | |
559 | } | |
560 | ||
561 | ||
562 | ############################################################################ | |
563 | # Module return value and documentation | |
564 | ############################################################################ | |
565 | ||
566 | 1; | |
567 | __END__ | |
568 | ||
569 | =head1 NAME | |
570 | ||
571 | Pod::PlainText - Convert POD data to formatted ASCII text | |
572 | ||
573 | =head1 SYNOPSIS | |
574 | ||
575 | use Pod::PlainText; | |
576 | my $parser = Pod::PlainText->new (sentence => 0, width => 78); | |
577 | ||
578 | # Read POD from STDIN and write to STDOUT. | |
579 | $parser->parse_from_filehandle; | |
580 | ||
581 | # Read POD from file.pod and write to file.txt. | |
582 | $parser->parse_from_file ('file.pod', 'file.txt'); | |
583 | ||
584 | =head1 DESCRIPTION | |
585 | ||
586 | Pod::PlainText is a module that can convert documentation in the POD format (the | |
587 | preferred language for documenting Perl) into formatted ASCII. It uses no | |
588 | special formatting controls or codes whatsoever, and its output is therefore | |
589 | suitable for nearly any device. | |
590 | ||
591 | As a derived class from Pod::Parser, Pod::PlainText supports the same methods and | |
592 | interfaces. See L<Pod::Parser> for all the details; briefly, one creates a | |
593 | new parser with C<Pod::PlainText-E<gt>new()> and then calls either | |
594 | parse_from_filehandle() or parse_from_file(). | |
595 | ||
596 | new() can take options, in the form of key/value pairs, that control the | |
597 | behavior of the parser. The currently recognized options are: | |
598 | ||
599 | =over 4 | |
600 | ||
601 | =item alt | |
602 | ||
603 | If set to a true value, selects an alternate output format that, among other | |
604 | things, uses a different heading style and marks C<=item> entries with a | |
605 | colon in the left margin. Defaults to false. | |
606 | ||
607 | =item indent | |
608 | ||
609 | The number of spaces to indent regular text, and the default indentation for | |
610 | C<=over> blocks. Defaults to 4. | |
611 | ||
612 | =item loose | |
613 | ||
614 | If set to a true value, a blank line is printed after a C<=head1> heading. | |
615 | If set to false (the default), no blank line is printed after C<=head1>, | |
616 | although one is still printed after C<=head2>. This is the default because | |
617 | it's the expected formatting for manual pages; if you're formatting | |
618 | arbitrary text documents, setting this to true may result in more pleasing | |
619 | output. | |
620 | ||
621 | =item sentence | |
622 | ||
623 | If set to a true value, Pod::PlainText will assume that each sentence ends in two | |
624 | spaces, and will try to preserve that spacing. If set to false, all | |
625 | consecutive whitespace in non-verbatim paragraphs is compressed into a | |
626 | single space. Defaults to true. | |
627 | ||
628 | =item width | |
629 | ||
630 | The column at which to wrap text on the right-hand side. Defaults to 76. | |
631 | ||
632 | =back | |
633 | ||
634 | The standard Pod::Parser method parse_from_filehandle() takes up to two | |
635 | arguments, the first being the file handle to read POD from and the second | |
636 | being the file handle to write the formatted output to. The first defaults | |
637 | to STDIN if not given, and the second defaults to STDOUT. The method | |
638 | parse_from_file() is almost identical, except that its two arguments are the | |
639 | input and output disk files instead. See L<Pod::Parser> for the specific | |
640 | details. | |
641 | ||
642 | =head1 DIAGNOSTICS | |
643 | ||
644 | =over 4 | |
645 | ||
646 | =item Bizarre space in item | |
647 | ||
648 | (W) Something has gone wrong in internal C<=item> processing. This message | |
649 | indicates a bug in Pod::PlainText; you should never see it. | |
650 | ||
651 | =item Can't open %s for reading: %s | |
652 | ||
653 | (F) Pod::PlainText was invoked via the compatibility mode pod2text() interface | |
654 | and the input file it was given could not be opened. | |
655 | ||
656 | =item Unknown escape: %s | |
657 | ||
658 | (W) The POD source contained an C<EE<lt>E<gt>> escape that Pod::PlainText didn't | |
659 | know about. | |
660 | ||
661 | =item Unknown sequence: %s | |
662 | ||
663 | (W) The POD source contained a non-standard internal sequence (something of | |
664 | the form C<XE<lt>E<gt>>) that Pod::PlainText didn't know about. | |
665 | ||
666 | =item Unmatched =back | |
667 | ||
668 | (W) Pod::PlainText encountered a C<=back> command that didn't correspond to an | |
669 | C<=over> command. | |
670 | ||
671 | =back | |
672 | ||
673 | =head1 RESTRICTIONS | |
674 | ||
675 | Embedded Ctrl-As (octal 001) in the input will be mapped to spaces on | |
676 | output, due to an internal implementation detail. | |
677 | ||
678 | =head1 NOTES | |
679 | ||
680 | This is a replacement for an earlier Pod::Text module written by Tom | |
681 | Christiansen. It has a revamped interface, since it now uses Pod::Parser, | |
682 | but an interface roughly compatible with the old Pod::Text::pod2text() | |
683 | function is still available. Please change to the new calling convention, | |
684 | though. | |
685 | ||
686 | The original Pod::Text contained code to do formatting via termcap | |
687 | sequences, although it wasn't turned on by default and it was problematic to | |
688 | get it to work at all. This rewrite doesn't even try to do that, but a | |
689 | subclass of it does. Look for L<Pod::Text::Termcap|Pod::Text::Termcap>. | |
690 | ||
691 | =head1 SEE ALSO | |
692 | ||
693 | L<Pod::Parser|Pod::Parser>, L<Pod::Text::Termcap|Pod::Text::Termcap>, | |
694 | pod2text(1) | |
695 | ||
696 | =head1 AUTHOR | |
697 | ||
698 | Please report bugs using L<http://rt.cpan.org>. | |
699 | ||
700 | Russ Allbery E<lt>rra@stanford.eduE<gt>, based I<very> heavily on the | |
701 | original Pod::Text by Tom Christiansen E<lt>tchrist@mox.perl.comE<gt> and | |
702 | its conversion to Pod::Parser by Brad Appleton | |
703 | E<lt>bradapp@enteract.comE<gt>. | |
704 | ||
705 | =cut |